Getting Started

Launching the TUI

Run models with no arguments to open the interactive terminal UI:

models

The TUI opens on the Models tab by default. You can change the default tab in your Configuration file.

Tabs

The TUI has four tabs, switched with [ (previous) and ] (next):

Tab	Description
Models	Browse AI models by provider, filter by capabilities, view pricing and metadata
Agents	Track AI coding assistants, view changelogs, check for updates
Benchmarks	Compare model performance with charts and head-to-head tables
Status	Monitor provider health with live incident and maintenance tracking

Global keybindings

These work in every tab:

Key	Action
[ / ]	Switch tabs
?	Show context-aware help popup
q	Quit

Navigation

All scrollable lists and panels share these keybindings:

Key	Action
Down / j	Move down
Up / k	Move up
g	Jump to first item
G	Jump to last item
Ctrl+d / PageDown	Page down
Ctrl+u / PageUp	Page up
Tab / Shift+Tab	Switch panel focus
Left / Right	Switch panels

Search

Most tabs support search:

Key	Action
/	Enter search mode
Enter / Esc	Exit search mode
Esc (in normal mode)	Clear current search

When searching, the query appears in the panel title as [/query]. The Agents tab also supports n / N to jump between matches.

Panel focus

Each tab has multiple panels (e.g., provider list, model list, detail panel). The focused panel has a cyan border; unfocused panels have a dark gray border. The selected item in the focused panel shows a > caret.

Use Tab or arrow keys to move focus between panels.

Help popup

Press ? in any tab to open a context-aware help popup showing all available keybindings for the current tab. Scroll with arrow keys and dismiss with ? or Esc.

CLI quick reference

Every TUI feature also has a CLI equivalent. Here are the most common commands:

# Models
models list                    # Interactive model picker
models show claude-opus-4-5    # Show model details
models providers               # List all providers

# Agents
agents status                  # Status table with versions
agents claude                  # Browse Claude Code releases
agents latest                  # Releases from the last 24h

# Benchmarks
benchmarks list                # Interactive benchmark picker
benchmarks show gpt-4o         # Show benchmark details

# Status
models status list             # Interactive provider health picker
models status show openai      # Detailed provider status
models status status           # Quick dashboard table

All commands support --json for scripting. See the individual feature pages for full CLI documentation.

Configuration

Settings are stored in ~/.config/models/config.toml. The file is created automatically when you customize tracked agents or providers. See Configuration for the full reference.

Next steps

• Models -- Learn about capability indicators, filters, and sorting
• Agents -- Set up agent tracking and version detection
• Benchmarks -- Compare models with charts and tables
• Status -- Monitor your AI providers
• Configuration -- Customize tracked agents, providers, and display settings